The present invention relates to targeted delivery, by means of intact killed bacterial cells, of bioactive molecules, including therapeutic nucleic acids, functional nucleic acids, drugs, peptides, proteins, carbohydrates and lipids, to mammalian host cells.
A number of hurdles continue to challenge targeted delivery of bioactive molecules to mammalian cells (e.g., cancer cells), particularly in-vivo. Those hurdles include (a) composition, functional characteristics and stability of delivery vehicles, (b) packaging therapeutically significant concentrations of bioactive molecules, (c) targeting desired diseased cells in-vivo, (d) overcoming a series of intracellular barriers and successfully deliver therapeutic concentrations of bioactive molecules to intracellular targets, (e) avoiding a range of host immune elements such as antibodies, complement, and macrophages that may destroy a vector before it reaches a target, (f) crossing the endothelial barrier of blood vessel walls, particularly at the site of a tumor mass, (g) migrating through several layers of cells to reach a target (e.g., it is known that a solid tumor is an organized structure containing both tumor cells and normal cells; hence a vector must cross several layers of normal cells to access malignant cells), (h) migrating through an extracellular matrix (ECM) comprised of glycoproteins, sulfated glycosaminoglycans, hyaluronan, proteoglycans and collagen that fills the space between cells and therefore hampers transport of a vector, and (i) addressing high interstitial hypertension (elevated hydrostatic pressure outside blood vessels) in the tumor microenvironment, which may limit the access of bioactive molecules.
A number of different vectors have been proposed for both nucleic acid and drug delivery, including viral, non-viral non-living, and non-viral living vectors. The non-viral non-living vectors have been adapted for both nucleic acid and drug delivery. The other two types of vectors have been adapted for nucleic acid delivery. Non-viral living vectors are mainly being developed for direct tumor-cell killing capabilities. While all these vectors have advantages, they also have drawbacks.
Viral vectors, such as retrovirus, adenovirus, adeno-associated virus, pox virus, herpes simplex virus, and lentivirus, have been developed for gene delivery. However, viral vectors are unable to deliver genes systemically and specifically to primary and/or metastasized tumor cells without infecting normal tissues (Akporiaye and Hersh, 1999; Biederer et al., 2002; Green & Seymour, 2002). Additionally, the extremely limited diffusibility of virions within extracellular spaces significantly hinders the dissemination of viral vectors. Moreover, viruses are antigenic, and therefore give rise to host immune responses. Such immune responses include both specific adaptive responses and non-specific innate responses (Chen et al., 2003; Ferrari et al., 2003; Wakimoto et al., 2003). The latter plays an important role in eliminating adenoviral vectors (Liu and Muruve, 2003) and HSV (Wakimoto et al., 2003).
Non-viral non-living vectors are exemplified by cationic polymers (polyplexes), cationic lipids (liposomes, lipoplexes) and synthetic nanoparticles (nanoplexes). They are more versatile than viral vectors, and offer several distinct advantages because their molecular composition can be controlled, manufacturing and analysis of such vectors is fairly simple, they can accommodate a range of transgene sizes (Kreiss et al., 1999; de Jong et al., 2001) and they are less immunogenic (Whitmore et al., 1999, 2001; Dow et al., 1999; Ruiz et al., 2001). The efficiency of gene delivery with non-viral non-living vectors is significantly less, however, than with viral vectors. At least 106 plasmid copies are needed to transfect a single cell, with approximately 102-104 copies actually making it to the nucleus for transgene expression (Felgner and Ringold, 1989; James and Giorgio, 2000; Tachibana et al., 2002). This inefficiency is attributable to the inability of non-viral non-living vectors to overcome the numerous challenges encountered between a site of administration and localization in a target cell nucleus, including, (a) the physical and chemical stability of DNA and its delivery vehicle in the extracellular space, (b) cellular uptake by endocytosis, (c) escape from the endosomal compartments prior to trafficking to lysosomes and cytosolic transport, and (d) nuclear localization of the plasmid for transcription. In addition to these physical and chemical obstacles, biological barriers, such as immunogenic responses to the vector itself and immune stimulation by certain DNA sequences containing a central unmethylated CpG motif exist (Yew et al., 1999; Scheule, 2000; Ruiz et al., 2001).
As an alternate to non-living nucleic acid/drug delivery vehicles, live bacterial vectors have also been developed for tumor targeted therapy (Pawalek et al., 2003; Soghomonyan et al., 2005). These vectors do not carry a payload of nucleic acids or drugs, but preferentially accumulate in tumor cells, replicate intracellularly and kill the infected cells (Pawelek et al., 1997). This phenomenon is thought to be facilitated by a complex bacterial system for introducing bacterial proteins directly into mammalian cells, which can result in the induction of apoptosis (Chen et al., 1996; Monack et al., 1996; Zhou et al., 2000). Currently, Bifidobacterium (Yazawa et al., 2000; 2001; Li et al., 2003), Clostridium (Minton et al., 1995; Fox et al., 1996; Lemmon et al., 1997; Theys et al., 2001; Dang et al., 2001; Nuyts et al., 2002a; 2002b; Liu et al., 2002) Salmonella (Pawelek et al., 1997; Low et al., 1999; Platt et al., 2000; Luo et al., 2001; Rosenberg et al., 2002) and Vibrio (Yu et al., 2004) are under investigation as tumor-selective live bacterial vectors.
Live attenuated bacteria have also been explored as vehicles for delivering nucleic acids (Paglia et al., 2000; Weiss and Chakraborty 2001; Yuhua et al., 2001), which may encode angiogenic inhibitors (Lee et al., 2005a; 2005b; Li et al., 2003), prodrug-converting enzymes (King et al., 2002) or cytokines (Yamada et al., 2000). Significant drawbacks of this approach include (a) live recombinant bacteria gradually lose plasmid DNA in vivo, mainly due to the absence of selection pressure and associated plasmid segregation, (b) bacteria carrying plasmid DNA tend to have a lower growth rate and appear to accumulate at lower levels and reside for a shorter period of time within tumors than bacteria without plasmids, (c) live Gram-negative bacterial vectors can cause severe endotoxin response in mammalian hosts, possibly due to in-vivo shedding of endotoxin (lipopolysaccharide; LPS), and evoke a Toll-like receptor response due to cellular invasion, (d) most of the tumor-targeting live bacteria accumulate and grow in the necrotic and relatively hypoxic foci within tumors, but not in well-oxygenated tumors at the rim of the growing nodules where tumor cells are normally most aggressive, (e) the risk associated with possible reversion to a virulent phenotype of these bacteria is a major concern (Dunham, 2002), and (f) the risk of infecting normal cells may lead to bacteremia and associated septic shock. The latter may particularly be a problem in immuno-compromised patients, such as late stage cancer patients.
Because problems continue to hamper the success of cancer therapeutics in particular, an urgent need exists for targeted delivery strategies that will either selectively deliver bioactive agents to tumor cells and target organs, or protect normal tissues from administered antineoplastic agents. Such strategies should improve the efficacy of treatment by increasing the therapeutic indexes of anticancer agents, while minimizing the risks of therapy-related toxicity.
The present invention provides a versatile delivery vehicle for improved drug, therapeutic nucleic acid and functional nucleic acid delivery strategies, especially but not exclusively in the context of cancer chemotherapy.